We’re happy to announce Rails 5 RC2 has been released.
We continue to work torwards the final Rails 5 release. Between RC1 and RC2 Rails got 91 commits! That’s a whole lot of bug fixes and polishing!
To view the changes for each gem, please read the changelogs on GitHub:
Hello everyone! This is Roque bringing latest news from the Rails community.
This week 23 people contributed to Rails. We also got 4 first time contributors. Welcome aboard folks and keep it going!
This fixes a scenario that caused incorrect template digests for */* requests that render non-HTML (e.g. JSON) templates. The HTML template was being used to compute the cache instead of the requested type.
FinderMethods#exists? from raising RangeErrorIt will return a boolean value instead of going 💥 when value is out range.
The checker will now play nice with multi-threaded web servers like Puma. Please read the pull request for more details.
If you have a custom queue adapter for Active Job, you can override the method queue_adapter_for_test in your tests to take advantage of the test helpers provided.
That’s it from This Week in Rails! There were many other great contributions, too numerous to list here, but feel free to check them out!
Until next week!
Ahoy hoy! Tim (aka Godfrey[9]) here bringing you all the latest from the Rails community. It’s been yet another killer week of activity as we continue to refine Rails 5 ahead of its upcoming launch 🚀. Don’t touch that dial - it’s gonna be a helluva show!
This week saw contributions from 26 fabulous people. 3 of those had a commit merged into Rails for the very first time. A big thank you to you all! ❤️
If you fancy seeing yourself up there next week, why not take a peek at the list of current issues? Improvements to the documentation can also be a great place to start!
Last week brought us an updated look at Rails by none other than DHH himself. The video offers a whirlwind tour of some of the most powerful features of the framework for newcomers in a tutorial that gets a blog set up in around 20 minutes. It also offers something for those looking to integrate newer features into their existing Ruby on Rails applications.
A recent regression caused #send_file to fail to halt the request cycle when used in a before_action. This is because the callback terminator checks for @_response_body , which is no longer set by #send_file. This fix updated to use #performed? instead.
If you use broadcasting to send messages to multiple loggers, you may have found that it does not silence all of them when told to do so. The fix updated to properly delegate #silence to all.
Kernel#caller_locations was introduced in MRI 2.0, and outperforms Kernel#caller in Module#delegate by around 10%. For the author this resulted in a faster boot time.
The new_framework_defaults file got a spruce this week, and documentation was updated to include better advice for upgrading to the new defaults from older apps.
The change also utilizes the update flag in several places to provide better messages catered for those only on the upgrade path.
That’s it from This Week in Rails! There were many other great contributions, too numerous to list here, but feel free to check them out!
Until next week!
Hey, team. Todd here, bringing you another edition of This Week in Rails! Due to a scheduling mishap, we missed an issue last week - we won’t cover it here, but I encourage you to check out the repo activity during that period.
Anyway, let’s get to the exciting stuff from this week!
First off, a very special thanks to Suhan Lee for translating our newsletter into Korean! He’s going to try to translate each newsletter going forward. If you’re interested in providing more translations, feel free to get in touch with Godfrey.
We had 25 of you lovely individuals contribute code and documentation to Rails this week. You all rock! Keep the contributions coming!
A late addition to Rails 5: default Rails initializers will no longer live in their own files. Instead, a new new_framework_defaults.rb file will be included in new apps generated with Rails 5.
WebSocket and logger Configuration Options for Action CableYou can now configure your own WebSocket and logger options when using Action Cable.
This patch decreases the size of model dumps to YAML - in some cases by a whopping 80%!
#exists? to #includesSean was able to fix an issue where chaining #exists? on #includes would raise an error. It sounds like there’s more work to do here yet, so I encourage giving the commit message a look-see.
ActiveRecord::Base#hash Should Differ Between ClassesPreviously, there was an issue where it was possible to have collisions if #hash was called on two different models with the same ID. This is now fixed on master.
This patch fixes deprecation warnings caused by the OpenSSL::Cipher::Cipher namespace changing to simply OpenSSL::Cipher.
That’s a wrap for This Week in Rails! There were plenty of other things going on, so feel free to jump in and check them out yourself!
See you next week!
Shhh, stay quiet! Rails 5.0 lurks just around the corner and it’s a mean beast.
Rails’ maintainers, computer scient(-ish)sts, spliced commit DNA merged long ago in hopes of creating a better web framework. The result is Jurrails Park.
Here’s Kasper alongside self-volunteering co-editor Jeff Goldblum, saying:
They were so preoccupied with whether or not they could, they didn’t stop to think if they should.
Contributors! Contributors! Contributors!
62 commits and 22 people makes us a happy newsletter team ❤️
The next version of Ruby, 2.4, unifies Fixnum and Bignum into Integer. Have no fear, Rails is already ready with both forward compatibility, backward compatibility and all around compatible compatibility. Int’ no Fix’ too Big’, folks.
Action Mailer can now rescue itself from exceptions with rescue_from.
This lets the mailer handle errors from the delivering job when run with Active Job.
Third, the exhaustively documented pull request changes how rescue_from handles Exception causes. Read more.
Calling suppress within another suppression on the same class was previously a let down.
Now the method doesn’t take its name too seriously and all should be good.
One of the first Rails 5.1 features brings the framework appropriately forward. Literally by forwarding count on a Relation to Enumerable when passed a block.
Rails has long had Time#all_day to really expand the horizon of past times. Now Date gets to go all_day too.
Though personally, dating for a whole day sounds pretty extreme, but kids these days ¯\_(ツ)_/¯
That concludes our report for this week. There were plenty of other things going on, so feel free to jump in and check them out yourself!
Live from RailsConf, happy 100th newsletter from Godfrey, Claudio, Vipul, Roque and Prathamesh (plus Kasper and Jon joining remotely).
The train from Kansas City is leaving the station soon. Just in time for a quick recap of what happened at RailsConf 2016 and this week in Rails.
This week 26 people contributed to Rails, including a first-time contributor. Thanks to your support, we are on our way to almost 5,000 all-time contributors!
Jeremy promised in his keynote that Rails 5 RC 1 will be released during the conference and it is here now!
Another key milestone towards the final Rails 5 release.
In the next weeks, Confreaks will be posting individual videos for each session. Meanwhile, you can still catch the full-day videos for Day 1, Day 2, and Day 3.
If you attended the RailsConf, don’t forget to let the organizers know your feedback before May 13th!
With this helper proxy, users can reuse helpers in the controller without having to include all the modules related to view context.
ActiveModel::RangeErrorWhen provided with large numbers, Active Model now fails with a new ActiveModel::RangeError that makes it easier to rescue from, and inherits from RangeError to maintain backward compatibility.
Rails session is now compatible with other Rack frameworks like Sinatra that are mounted in Rails. They can also use session tooling of Rails without any issues now.
Meeting you all at RailsConf was a lot of fun. And if you haven’t seen us up on stage during the lightning talks, here is the YouTube link of “Goodfry and his boy band”.
That concludes our report for this week. There were plenty of other things going on, so feel free to jump in and check them out yourself. Also look out for the announcement of all the videos from RailsConf on Confreaks!
See you next year at RailsConf 2017 in Phoenix AZ!
Dear Mayor,
Welcome to Kansas City! This is your advisor Godfrey. Since you are new here, it must be a bit overwhelming for you. Don’t worry, I’m here to help you do your job. Let’s get to it!
Rails 5.0.0.beta4 was released this week. Based on the increasing version numbers, I speculate the core team is working towards a 5.0.0.RC1 release and eventually a 5.0.0 final release. I recommend you try out the beta release on our apps.
According to sources close to the development, 34 contributors were seen submitting patches to the Rails code base this week. From what I can tell, this appears to be a good sign.
Our plans to import colors from Oz have worked – I am proud to say Kansas is no longer black and white. I recommend that you install the Guidebook app ahead of the conference. You might also want to sign up for the workshops too – see the official welcome email in your inbox for details.
We are not interested in recycling the dishwater or taking 30-second showers. Worse than that, some of the sponsors have planned parties for the attendees. How can we call ourselves a decent city if we can’t accommodate that?
Put some drink in your think, Mayor, and get us some water.
Despite already accommodating well over a thousand attendees, our polls have shown the majority of Rails users will not be able to attend the conference.
To prevent a public outcry, I advised the conference organizers to record all talks and make them available on this website after the conference.
They are planning to live stream the keynotes as well, although I heard the secret URL won’t be up until next Wednesday. I can hear the sound of F5 buttons cracking up from here.
In a series of experiments to research Swearing as a response to pain, a local scientist has discovered some new methods to manipulate time by accident. As a result of this new discovery, these core extension methods will be available in Active Support 5.
Also new to Active Support, Rails users will soon be able to express themselves in these newfound vocabularies.
A scientific study to end the decade-old debate in the programming community has concluded in a surprising twist. The report declared that faster code does seem to perform better, after all.
As a follow up to the research, this contributor was the first to apply the new optimization technique to Array#split. The work has resulted in some performance gains, contrary to everyone’s expectations.
A group of linguists gathered in town to discuss the merits of semicolons. During a panel discussion, they have reached consensus that typos can indeed reduce readability.
Coincidentally, this week the Rails Guides have received a flurry of changes that improved their readability.
My recommendation? We should sit down and fix the typos in our city bylaws some day.
It is well-known in the computer science community that concurrency is hard. This patch that just landed on Rails master is the perfect example to illustrate why – in an attempt to fix a bug in Active Support’s ShareLock class, the patch was briefly misidentified as Sherlock Holmes, causing a brief delay and a race condition.
Recently, an audit discovered that Rails sometimes loads the database schema twice. According to a survey, most Rails users do not find this necessary, leading to the “feature” being removed from master this week.
I have an unfortunate piece of news for you, Mayor. The JRuby builds have been temporarily flagged as allowed to fail on Rails’ CI servers. If you know anyone who works with JRuby, we could use some help identifying and fixing the root causes here.
Despite promising results from last week, the researchers involved have backtracked their claims on the performance benefits associated with this change. Based on my negative lookahead, I believe this issue will remain a source of huge debate among the experts in the foreseeable future.
That concludes my report for this week. There were plenty of other things going on, so feel free to jump in and check them out yourself.
See you next week at RailsConf!
We’re happy to announce Rails 5 Beta 4 release. We’re down to 4 open issues on the issue tracker for Rails 5.0.0.RC1.
We continue to work torwards RC1. Between beta3 and beta4 Rails got over 1,000 commits!
To view the changes for each gem, please read the changelogs on GitHub:
Hello everyone! Prathamesh here bringing you This Week in Rails. Let’s get started.
Thank you to 32 awesome people who contributed to Rails this week, including 5 first-timers! Want to join the party? Get started with the issues page.
String#blank? just became faster 🏇This speed optimization will improve the performance of blank? calls against empty strings by about 3.5x. It was further improved to about 30% faster!! BOOM 💪
It is possible to specify comments for tables, columns, and indexes in the database itself now with this addition. It currently works for MySQL and PostgreSQL adapters.
create_join_table works with non-integer column typesCreating a join table with create_join_table helper used to always create the columns with integer type. But now if you want to have uuid columns or any other type, it’s possible!
This fixed issues with intermittent test failures related to the Active Record connection pool trying to return connections at the wrong time.
Rails recently improved performance for inserting schema information using multi-row-insert. But some SQLite versions do not support the multi-row-insert feature which resulted into errors. Now Rails will gracefully handle this situation without an error.
That’s all for This Week in Rails. There were many more improvements than we have room to cover here, so don’t hesitate to jump in and check them out yourself!
See you next week!
Hello everyone!
Jon here bringing you This Week in Rails from the sunny surf town of Biarritz, France 🇫🇷☀️
Ready to jump in and catch a few waves of updates?
Grab your surfboard and let’s go!
Please share some aloha 🌸 for the 27 daring riders who contributed to Rails (and world happiness) this week, including 4 committers who caught their first wave 😍😎! You can, too! Check out the current issues if you’d like to paddle out, make new friends and join the fun 💙
Thanks to Jeremy Daer, Ryuta Kamizono, Vipul A M and Ian Gilfillan, Active Record is now fully compatible with MariaDB ❤️💛💚💜… click on our heroes’ names to open the commits and see what went down behind the scenes!
MariaDB is a community-developed open source fork of MySQL. Notable users include Wikipedia, Facebook, Google and, yes, Basecamp 3! Here is a good Wikipedia article about MariaDB and MySQL history. Game on!
Everybody out of the pool! 🏊 It was getting crowded in the water, but luckily our experienced Baywatch lifeguards Matthew, Sean and Jeremy 🏋 were on hand to, er, pool their efforts to ensure your database connections are managed automatically when Action Cable is involved! Synchronized water ballet, anyone?
No Strings are better than NoSQL, right? Ankit noticed that the number_to_delimited and number_to_rounded helper methods were creating duplicate strings. Check out the pull request to see how he improved it using derailed_benchmarks to check object allocations 🍰
max_time edge casesRails has the need for speed 💨 and surfers Blake and Xavier managed a crowd-pleasing acceleration maneuver in AS::FileUpdateChecker using a single loop memoization rather than multiple loops, benchmarked using benchmark-ips. It’s good stuff; see the PR for more! 🍦
attribute_changed? in ActiveModel::DirtyFaster than relocating to San Francisco ✈️ (and much more fun), sun-drenched rider Lihan was able to speed up attribute_changed? by setting the default options parameter to nil instead of a hash, and checking the result with benchmark-ips 😎
Lifeguard Jeremy came across a common, hard-to-detect 🐞 in the water: Local message changes weren’t being enqueued with the delivery job when using #deliver_later. This now raises a friendly exception to make sure your message modifications are made in either (a) the mailer method or (b) using a custom Active Job. Nice catch! 🌊
While surfing Mongoid’s test suite, an error alerted our champion Emily 🏄🏾 to a change of behavior in ActiveSupport#symbolize_keys: Rails 4.1 returned a Hash but 4.2/5.0 returned self.class. This launched an interesting discussion on the merits of expected behavior versus matching Ruby’s behavior. Have a look! 🍎
That’s all for This Week on the Beach with Rails™ 🌴. There were many more improvements (and great waves) than we have room to cover here, so don’t hesitate to jump in and check them out yourself… the water’s warm and the locals are friendly! 👣🐾
See you next week! 🍍🍌🍉🍇